Ciklum Speaker’s Corner – Building full stack AI chatbots with Java (Ciklum, 2023)

Info

Project type Public Speaking, Tech Presentation
Date 22nd of Feb 2023
My Role Speaker
Registered 205, from 9 countries
Topics AI, Java, Chatbots, NLP, NLU, Machine Learning
Keywords Conversational AI, Natural Language Processing and Understanding, Telegram Bots
Skills developed Presentation, Speaking
Event URL Ciklum's website
Event recording (Youtube)

Description

This was my first live coding session in front of an international audience of ~40 professionals (devs, devops, managers and product owners). I had the great opportunity to demonstrate how to build a full-stack AI chatbot with Java, in a ~2h live demo session at Cikulm speaker's corner.

It was a challenging experience, as preparations for such events are demanding. I learned from this that writing the code and setting up the environment is not the complicated part - but handling the exceptions and new scenarios during the demo is.

Thanks for the opportunity Iryna Rud and thanks everyone for your feedback.

This session is based on the experienced I gained in one of the production projects that I worked on for Flatirons and on the knowledge that I gained during the preparation for my AI Developer certification from GTC, in 2022. While preparing for the certification exam, I was also working to publish a series of articles that helped me learn the fundamentals easier and have them organized in my mind.

My purpose in this session was to explain and demonstrate how one can build a conversational chatbot by only writing code (no subscription, no low-code platform needed).

This event came with a lot of challenges that were new to me, such as:

presenting in front of people around the Globe, to an unknown public
presenting to a heterogeneous group of people with such different levels of expertise and having a wide spectrum of skill sets
representing publicly such as strong brand as Ciklum is
being broadcasted and recorded with public access without the benefit of doubles

It was an experience for which I prepared myself intensively and worked hard to overcome my own limits. I also learned a lot of new things and gathered valuable feedback.

I am very grateful to Ciklum's branding team for the invitation and for the opportunity. Thanks!

References

Github repo
Presentation slides
Architecture diagram
Video recording (raw)
Video recording (Ciklum's YouTube channel)
My articles related to this topic

Tech stack

- Java, openjdk-19, maven 3.8.1
- Spring
- Stanford CoreNLP, Natural Language Processing Java lib
- Lingua, Language detector Java lib
- Jsoup, a powerful HTML parser for Java
- Telegram Bots API doc
- Telegram Bot, Java lib
- Docker
- Env setup commands
- Solr, powerful NoSQL DB & search engine
- MariaDb, a powerful open-source RDBMS
- Metabase, an amazing Java open-source tool for analytics